OpenAI unveils Codex, an advanced AI coding agent

Codex is currently available to all the subscribers to ChatGPT Pro, Enterprise, and Team

OpenAI unveils Codex, an advanced AI coding agent
OpenAI unveils Codex, an advanced AI coding agent

OpenAI officially launched an advanced cloud-based software engineering agent called Codex.

Codex, OpenAI’s most capable AI coding agent, is powered by Codex-1. According to OpenAI, Codex-1 generates “cleaner” code than o3, offering enhanced accuracy and will repeatedly run tests on its code until passing results are successfully accomplished.

OpenAI Codex consists of enhanced communication capabilities

The ChatGPT manufacturer stated that the AI coding agent is capable of communicating anywhere from one to 30 minutes to resolve queries, write simple features, fix bugs regarding your codebase, and perform a variety of tasks.

In addition, the recently launched agent can manage several software engineering tasks simultaneously without restricting users to access their computer and browser while it’s running.

Availability

Codex is currently available to all the subscribers to ChatGPT Pro, Enterprise, and Team.

Notably, the company is set to offer “generous access” to Codex to start soon, with further plans to expand Codex access to ChatGPT Plus and Edu users in the near future.

It is pertinent to mention that OpenAI is bringing a significant update to Codex CLI, the latest open-source coding agent, with a variant of its o4-mini model.

To note, this model is now the default in Codex CLI, and will be accessible in OpenAI’s API for $1.50 per 1M input tokens and $6 per 1M output tokens.

Read more : Sci-Tech
Google introduces YouTube mobile video editor to iOS, taking on CapCut
Google introduces YouTube mobile video editor to iOS, taking on CapCut
YouTube Create received 500,000 Android downloads in Q2 2025, with less than 4 million since launch
OpenAI takes major step to reduce Nvidia chips dependency
OpenAI takes major step to reduce Nvidia chips dependency
Google helps OpenAI with the stategic shift from Nvidia's AI chips in major shift
iPhone 17 to feature an enlarged display than iPhone 16: Report
iPhone 17 to feature an enlarged display than iPhone 16: Report
Apple’s iPhone 17 and iPhone 17 Air are likely to be launched with Samsung's latest M14 OLED panels
Denmark takes major step to crack down on AI deepfakes: Gives citizens copyrights
Denmark takes major step to crack down on AI deepfakes: Gives citizens copyrights
Danish government to grant its citizens copyright over personal features to combat deepfakes
Gemini Scheduled Actions rolls out to Android, iOS, and web
Gemini Scheduled Actions rolls out to Android, iOS, and web
Gemini Scheduled Actions now available to Google AI Pro, AI Ultra subscribers or eligible Google Workspace plan
TikTok experiments it's own variant of Instagram’s ‘broadcast channels’
TikTok experiments it's own variant of Instagram’s ‘broadcast channels’
With this feature, TikTok aims to enable creators and brands to share updates and behind-the-scenes content
Meta negotiates acquisition of voice cloning startup Play AI: Report
Meta negotiates acquisition of voice cloning startup Play AI: Report
This move follows Meta CEO Mark Zuckerberg’s wider push to expedite Meta’s AI strategy
Google Photos brings significant update to ‘Ask Photos’ feature with AI
Google Photos brings significant update to ‘Ask Photos’ feature with AI
Google also announced the roll out of the best of Photos’ classic search feature into Ask Photos
Google releases Doppl app for AI-powered outfit try-ons
Google releases Doppl app for AI-powered outfit try-ons
Doppl app is currently accessible on iOS and Android across the US
YouTube introduces AI Overviews-like search results feature
YouTube introduces AI Overviews-like search results feature
YouTube's assistant-like feature offers video summaries, content suggestions, and academic quizzes,
Instagram brings custom colours and emojis to notes
Instagram brings custom colours and emojis to notes
Meta-owned Instagram’s latest update is currently available to users worldwide
Google introduces Gemini CLI to developers’ terminals for free
Google introduces Gemini CLI to developers’ terminals for free
Gemini CLI offers a usage limit of up to 60 model requests per minute and 1,000 requests on per day basis